The irreducible epistemic atoms underlying the curriculum. 4,828 atoms across 8 types and 2 molecules
Create explicit transactive memory systems by documenting who knows what, maintaining this meta-knowledge actively, and routing questions to the appropriate expertise holder.
Use pair programming and knowledge-sharing rotations to build redundancy into team transactive memory, creating multiple paths to critical expertise.
Engineer confirmed success pattern conditions into new projects from the start rather than hoping they emerge organically.
Begin each retrospective by reviewing previous action items and requiring explicit accountability—either completion or a decision to deprioritize—to prevent insight without implementation.
Structure retrospectives around comparing expectations to actual outcomes rather than just listing what happened, to reveal where team mental models were inaccurate.
Explicitly establish and enforce team norms that distinguish task conflict from relationship conflict, celebrating disagreement about ideas while immediately addressing personal attacks.
Design meetings with pre-work that moves cognitive effort out of synchronous time, allowing the meeting to focus on work that requires real-time interaction rather than individual processing.
Limit meeting attendance to only those who need to participate, because each additional person increases coordination costs and reduces individual accountability.
Design asynchronous collaboration with bounded comment periods, clear reviewers, and explicit decision mechanisms to prevent open-ended discussion that never concludes.
Separate high-signal channels requiring attention from reference channels consulted when needed and low-signal optional channels, because collapsing these into a single stream destroys the signal-to-noise ratio.
Design information routing with redundant paths for critical information, because single-path routing creates vulnerability to individual absence or channel failure.
Push urgent information to recipients through notifications while making reference information available for pull when needed, to prevent notification floods that teach people to ignore all notifications.
Allocate team attention explicitly across categories (committed priorities, responsive buffer, exploration) and make this budget visible to stakeholders to prevent reactive work from consuming all capacity.
Identify your default role in recurring relational patterns, then run controlled experiments changing one variable to test whether the pattern is as rigid as it feels.
Establish interrupt protocols that route different priority levels through different channels, preventing low-priority interruptions from receiving high-priority attention and distributing the cognitive cost of responsiveness.
Conduct attention retrospectives to review how the team actually spent attention versus intentions, creating a feedback loop for calibrating and improving attention management over time.
Attach new epistemic practices to existing team rituals rather than creating separate meetings to ensure adoption through reduced friction.
Develop individual epistemic skills as the foundation for team cognitive architecture, since team practices cannot compensate for low individual competence.
Individual metacognitive transparency—articulating reasoning, evidence, and uncertainties—enables teams to evaluate and build on thinking rather than just conclusions.
Organizational schemas should be made explicit through independent elicitation across roles and levels, with convergences, divergences, and absences each providing distinct diagnostic information about alignment, conflict, and blind spots.
Periodically surface process schemas by extracting embedded assumptions (about risk, capability, sequencing, quality) and evaluating them against current reality, because processes designed for past contexts become organizational fossils when their assumptions are no longer valid.
Conduct structured newcomer debriefs at 30/60/90 days to capture schema observations before habituation renders them invisible, because newcomers occupy a unique vantage point where organizational assumptions are still visible as choices rather than facts.
Frame strategy as a single causal model ('We win by doing X because Y') rather than as a list of objectives, because a causal model enables distributed coherent decision-making while a list only enables plan-following when the plan applies.
Design strategy statements to make tradeoffs explicit through 'prioritize X over Y' framing, because strategy's power comes from what it excludes and ambiguous priorities provide no guidance when values conflict.